Article Text

Download PDFPDF

Triple tracer (blue dye, indocyanine green, and Tc99) compared to double tracer (indocyanine green and Tc99) for sentinel lymph node detection in endometrial cancer: a prospective study with random assignment
  1. Roy Kessous1,
  2. Jeffrey How2,
  3. Jeremie Abitbol2,
  4. Sanam Puzhakkal2,
  5. Liron Kogan2,
  6. Amber Yasmeen2,
  7. Shannon Salvador2,
  8. Walter H Gotlieb2 and
  9. Susie Lau2
  1. 1 Department of Obstetrics and Gynecology; Faculty of Health Sciences, Soroka University Medical Center; Ben-Gurion University of the Negev, Beer-Sheva, Israel
  2. 2 Division of Gynecologic Oncology, Jewish General Hospital - McGill University Faculty of Medicine, Montreal, Québec, Canada
  1. Correspondence to Dr Walter H Gotlieb, McGill University Jewish General Hospital, Montreal, QC H3T1E0, Canada; walter.gotlieb{at}


Objective Sentinel lymph node (SLN) mapping is increasingly being used in the treatment of apparent early-stage endometrial cancer. The aim of this study was to evaluate whether three tracers (blue dye, indocyanine green (ICG), and technetium-99 (Tc99)) performed better than two (ICG and Tc99).

Study Design Prospective study of all consecutive patients (n=163) diagnosed with clinical early-stage endometrial cancer from 2015 to 2017. All patients were randomly assigned to receive a mixture of ICG and Tc99 with or without blue dye. Subgroup analysis for detection rates was performed for each group (double versus triple tracer).

Results One hundred and fifty-seven patients met the inclusion criteria. Eighty patients received ICG and Tc99 with unilateral and bilateral SLN detection rates of 97.5% and 81.3%, respectively. Seventy-seven patients received all three tracers with unilateral and bilateral detection rates of 93.5% and 80.5%, respectively. Only one patient in the triple tracer group was detected by blue dye alone. No significant differences were noticed in unilateral or bilateral detection rates between the two groups, nor in the detection of lymph node metastasis.

Conclusion The addition of blue dye to ICG and Tc99 did not demonstrate any improvement in SLN detection.

  • endometrial cancer
  • sentinel lymph node mapping
  • indocyanine green
  • blue dye
  • technetium

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Use of sentinel lymph node (SLN) mapping is increasing and defining the optimal technique is essential.

  • This prospective randomized study shows that adding blue dye did not improve the detection rate of SLN.

  • Removing blue dyes reduces the discoloration of tissues and the risk of allergic reactions.


Endometrial cancer is the most common gynecological cancer, with a rising incidence.1 Fortunately, the majority (70%) of endometrial cancers are diagnosed at an early stage, with good 5- and 10-year survival rates.2

Controversy remains concerning the optimal lymph node assessment tool, with approaches ranging from full lymph node dissection, selective nodal dissection (based on uterine factors), or no lymph node dissection.3 4 Sentinel lymph node (SLN) mapping has emerged as an attractive alternative to allow integration of information of lymph node status in the treatment algorithm,5 avoiding most of the morbidity of full pelvic lymph node dissection.

SLN mapping requires the injection of radiotracer and/or colored dye in order to locate the draining lymph node.6 Increasing evidence in the literature regarding SLN mapping demonstrates high accuracy, with 96% sensitivity for detection of metastatic disease and a negative predictive value of 99.7%.7

With regard to tracer options, blue dyes (methylene or patent) were the most commonly used, and indocyanine green (ICG) has become a more attractive alternative in recent years.8–10 With regard to toxicity of the three tracers, a relatively higher side effect profile is reported with the use of blue dyes compared with technetium-99 (Tc99) and ICG (2% vs 0.006% and 0.05%, respectively).10 Since its introduction, several studies have evaluated the use of ICG alone or in combination with blue dye and Tc99. ICG is associated with higher overall and bilateral detection rates compared with blue dye; however, no differences were found between ICG alone and the combination of blue dyes and Tc99.9 11

When using the three dyes together, we previously observed a bilateral detection rate of 76%, but the blue dye had lower bilateral detection rates (43%), compared with either ICG (65%) or Tc99 (71%) alone.9 We therefore decided to evaluate whether blue dye added any value and improved the detection rates by prospectively randomly assigning patients to the triple dye or to a combination of ICG and Tc99 without blue dye.


Study population and design

From June 2015 to May 2017, all consecutive patients with clinical stage I endometrial cancer were enrolled into the study. Patients who had apparent metastatic disease prior to surgical staging were not included. The study was approved by the hospital institutional research board (Protocol #2019–1536). All patients underwent robotic surgical procedures performed by one of three gynecologic oncologists (WHG, SL, SS). Prior to the initiation of this study, this gynecologic oncology team had performed more than 1000 robotic surgeries and had performed sentinel node sampling in more than 300 endometrial cancer cases in addition to experience with SLN mapping in cervical and vulvar cancers since 2003. All patients underwent total hysterectomy, bilateral salpingo-oophorectomy, and SLN dissection. In cases that a SLN was not detected, ipsilateral full lymph node dissection was performed in accordance with the SLN algorithm outlined by Memorial-Sloan Kettering12 and National Comprehensive Cancer Network (NCCN) guidelines.3

Data for the study were collected prospectively and maintained in a digital database. For each patient, we extracted information regarding age, body mass index (BMI), pre-operative and post-operative histologic type and grade, International Federation of Gynecology and Obstetrics (FIGO) stage, myometrial invasion, and the presence of lymphovascular space invasion (LVSI).

During the study period, we used:

  1. Patent blue (patent blue sodium injection Guebert product imported by Methapharm Inc., Brantford, ON, Canada)

  2. ICG (IC-Green, Akorn Pharmaceuticals, Lake Forest, IL, USA)

  3. Tc99 (99mTc-SC, microsulfur colloid, Pharmalogic PET Services, Montreal, Canada).

All enrolled patients were randomly assigned to receive a mixture of either all three tracers if they underwent surgery on uneven days (prepared in four 1 mL syringes, each containing a mixture of blue dye (0.8 mL), ICG (0.1 mL, 0.25 mg/mL), and Tc99 (0.1 mL)) or two tracers (ICG (0.1 mL, 0.25 mg/mL), Tc99 (0.1 mL), and normal saline (0.8 mL)) on even days. Using even and uneven days significantly facilitated implementation and consistency, and because surgeons operate on a fixed weekday, changing between even and uneven days changed the odds for a particular surgeon every week, that is, surgeon A would operate on Tuesday March 1 and patients would receive three dyes, and Tuesday March 8 and patients would receive two dyes, and so on.

Tracers were injected directly into the cervix at the 3 and 9 o'clock positions.13 14 Injection was performed following prepping of the patient under anesthesia and just prior to skin incision. At each of these positions, a 1 mL syringe was injected superficially (2–3 mm) into the cervical submucosa and another 1 mL was injected deep (3–4 cm) into the stroma of the cervix towards the lower uterine segment.

Detection of SLNs was accomplished through any of the three following methods:

  1. Direct visualization of blue colored lymphatics/nodes

  2. Visualization of green-colored lymphatics/nodes via the immunofluorescent imaging mode on the da Vinci surgical platform

  3. Detection of radioactive nodes by a handheld gamma probe (Daniel Probe, RMD Instruments Corp., Watertown, MA, USA). The reading of the counter was performed ex-vivo intra-operatively in order to serve as an additional confirmation for the detection of the SLN. In rare circumstances it allowed selection of the ‘hottest’ node when there was a smear of blue or green on the nodes.

The regional location of the SLN was documented by the nurses, as well as whether they were only blue, only radioactive, only fluorescent green, or any combination of the three. During all cases, a uniform electronic data sheet was utilized for documentation of the SLN location and the tracer(s) that detected the SLN.


All pre-operative biopsies were reviewed. All SLN samples were bisected and stained with hematoxylin & eosin (H&E). In our institution, routine ultra-staging was started in September 2016 and was performed in 73 patients (45%) of the studied population. Immunostaining for cytokeratin (clone AE1/AE3, Milipore Inc., dilution 1:150) was performed in SLNs after H&E histological examination. Non-SLNs were processed using entire node examination with H&E staining. Macrometastases were considered as tumor foci larger than 2 mm, micrometastases were considered as tumor foci between 0.2 and 2 mm, and isolated tumor cells were defined as deposits less than 0.2 mm.

Statistical analysis

Detection rate was calculated as the number of patients with at least one detected SLN divided by the total number of patients who underwent lymphatic staging. Power analysis revealed that a sample size of more than 60 patients in each group would be required to detect a difference of more than 5% in detection rates between the groups, with 90% power and an alpha of 0.05. Subgroup analysis for detection rates was done for each group (two versus three tracers). Given that the aim of the study was to evaluate whether adding blue dye is improving SLN technique on both unilateral and bilateral detection rates, we performed the analysis for SLN detection per patient and not per hemi-pelvis. Statistical analysis was performed using STATA 13 (StataCorp, College Station, TX, USA). Statistical significance was calculated using the Chi square test for differences in qualitative variables and the Student t-test for differences in continuous variables.


During the study period, 163 patients were enrolled in the study. Table 1 summarizes the general characteristics of the patients in the two groups. The groups did not differ significantly in any of the clinical characteristics including age, BMI, stage, histology, grade, myometrial invasion, and LVSI.

Table 1

Clinical characteristics of the two patient groups

Figure 1 presents an overview of the data. Six patients were excluded from the analysis; two patients did not have ICG and two patients did not have Tc99 due to lack of availability on the morning of surgery, one patient had a final diagnosis of primary high-grade serous carcinoma of the ovary, and one patient had a benign diagnosis on final pathology. Eighty patients received the double tracer mixture with unilateral SLN detection rate of 97.5% (78 patients) and bilateral SLN detection rate of 81.25% (65 patients). Seventy-seven patients received triple tracer mixture with unilateral SLN detection rate of 93.5% (72 patients) and bilateral SLN detection rate of 80.5% (62 patients). These differences were not statistically different (Table 2).

Figure 1

Organogram showing lymph node detection rates in the study population. SLN, sentinel lymph node; Tc99, technetium-99.

Table 2

Detection rate of sentinel lymph node mapping in the two patient groups

Figure 2A shows the SLN detection rate (bilateral and unilateral) according to anatomic location and comparison groups (blue versus green). When focusing on lymph node detection in the triple tracer group (n=72), looking at the tracer that guided the detection of the SLN, in 15 (21%) patients the SLN was detected by ICG alone and in just one patient the SLN was detected only by the blue color (Figure 2B).

Figure 2

(A) Sentinel lymph node (SLN) detection rate (bilateral and unilateral) according to anatomic location and comparison groups (blue versus green). (B) Within the triple tracer group (72 patients that had a SLN detected), detection of lymph nodes according to the type of tracer (green, blue, and both). Pt, patient.

Overall there were 17 patients in whom a metastasis was found in the lymph nodes (10.8%); 11 in the double tracer group and six in the triple tracer group (p=0.230). Of the 17 patients with metastatic disease in the lymph nodes, 12 had macro-metastases and five patients had micro-metastases. Of the 17 patients with metastatic disease in lymph nodes, SLNs were detected in 13 patients and in four patients full pelvic lymph nonde dissection was performed. Among the 13 patients with metastatic disease in which a SLN was detected, the positive node was the SLN in 11 cases. Two patients had SLNs detected in one hemi-pelvis and failed to map on the contralateral hemi-pelvis (one in each dye group), and full pelvic lymph node dissection was performed at the surgeon's discretion due to high-grade pre-operative histology and metastatic non-SLNs were detected.

When focusing on the group of patients with SLN detection (n=150), nine metastatic SLNs were detected in the double dye group (n=78) versus two in the triple dye group (n=72) (p=0.04) and two patients had non-SLN metastasis detected in the contralateral hemi-pelvis where SLNs were not detected.

Of the 157 patients in the study, seven had no SLNs detected and full pelvic lymph node dissection was performed. Of these seven patients, three had grade 1 disease and no lymph node metastasis was found in final pathology. The remaining four patients had metastasis to lymph node detected on final pathology; all four patients had grade 2–3 disease, in one patient after full pelvic lymph node dissection was performed one of the lymph nodes was Tc99-positive, two patients had enlarged pelvic lymphnodes and underwent full pelvic lymph node dissection, and one patient had extensive pelvic and abdominal disease spread observed during surgery.

During the study period, no toxic reactions or allergic reactions were reported in any of the patients.


Based on available data, the NCCN guidelines3 offer SLN protocol as an alternative to complete lymphadenectomy in patients with early-stage endometrial cancer. Given the increasing use of SLN mapping for these cases, it is important to define the optimal technique for this procedure, including choosing the most accurate tracer or combination of tracers. In this prospective study with random assignment, we demonstrate that the addition blue dye does not contribute to improved detection rates of SLN, if ICG or Tc99 are utilized.

In the quest to find the optimal tracer, initially most studies reported the use of a combination of blue dye (methylene or patent) and Tc99,15 with recent studies evaluating the use of ICG.7 9 16 ICG use quickly established itself because of high detection rates, high sensitivity and negative predictive value, and finally a reduced potential of complications rates.8 10 The use of Tc99 with blue dye was previously compared with ICG alone, and comparable unilateral detection rates were found between the two groups with significantly higher bilateral detection rates in favor of the ICG group.17 Detection rates as well as negative predictive value in our study compare well with recently published studies reporting detection rates of between 85% and 100%.16 18

In a recent randomized study by Frumovitz et al, 19 authors evaluated mapping substances and whether ICG is non-inferior to blue dye for the detection of SLNs in women with cervical and uterine cancers. Authors concluded that ICG identified significantly more SLNs than blue dye, with no difference in the pathological confirmation of the nodes. Based on a retrospective study on 100 patients that was published by our group using all three tracers (Tc99 + ICG + blue dye), it seemed that blue dye might not contribute much to the detection rates of SLNs.9 In the current study we have addressed this question in a prospective manner using random assignment, and show that the addition of blue dye did not improve either the detection rate or the accuracy of SLN mapping. In addition, based on randomization leading to two groups with similar patient characteristics, we unexpectedly found that the metastatic pick-up with triple dye was lower than with double dye (six versus 11 patients), although this did not reach statistical significance (p=0.23). It remains possible that the addition of blue interferes with migration of dye, mainly when the lymphatic flow might be hindered by metastases in the lymph node. Therefore, given the discoloration of tissues making dissection more challenging, and the potential for allergic reactions with blue dyes, these data provides evidence that one can use only ICG and Tc99 while maintaining high detection rates. Given that the technique for ICG use has several advantages compared with Tc99, the obvious next step is to evaluate the added value of Tc99 by comparing ICG alone to ICG with Tc99, and this prospective randomized study has presently been initiated.

Compared with the use of 'high-dose' ICG (5 mg/mL and a volume of 8 mL) that was recently published,20 we used four syringes, each with a mixture that included ICG 0.1 mL (0.25 mg/mL). This allowed similar bilateral detection rates to the reported high-dose injection group and have the advantage of being able to identify the SLN, rather than multiple nodes.

In our institution, we have been preforming SLN mapping since 2003, and more specifically for endometrial cancer since 2010. This study offers the advantage of a prospective randomized study with a considerate sample size and three experienced surgeons, leading to minimal variability in surgical techniques. However, the study is not without limitations. One limitation lies in our recent practice of ultra-staging, making us unable to comment on the management of cases with low-volume metastases. Another potential limitation lies in the randomization method that was used (by days of surgery) that does not completely prevents selection bias. Nevertheless, we have addressed this limitation by assigning patients for surgery day using administrative consideration (type of surgery, operating room time, and more) without considering patient characteristics that might have an effect on the study results. In addition, as presented in the Results section, the studied populations did not differ in those variables that might influence SLN technique or effectivity (such as obesity, age, and tumor characteristics).

In conclusion, in the search for the optimal technique for performing SLN mapping in early-stage endometrial cancer, the results of our study show that adding blue dye does not improve any of the detection parameters.


This work was made possible in part by grants from the Montreal-Israel Cancer Research Fund, the Gloria Shapiro Fund, the Anne Marie and Mitch Garber Fund, and the Susan and Jon Wener Fund.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.


  • Contributors RK: analysis and interpretation of data, drafting the article and revising it for submission, and approval of the version to be submitted. JH: acquisition of data, drafting the article and revising it for submission, and approval of the version to be submitted. JA: analysis and interpretation of data, revising the article for submission, and approval of the version to be submitted. SP: acquisition of data, revising the article for submission, and approval of the version to be submitted. LK: interpretation of data, revising the article for submission, and approval of the version to be submitted. AY: interpretation of data, revising the article for submission, and approval of the version to be submitted. SS: conception and design, acquisition of data, interpretation of data, revising the article for submission, and approval of the version to be submitted. WHG: conception and design, acquisition of data, interpretation of data, drafting the article and revising it for submission, and approval of the version to be submitted. SL: conception and design, acquisition of data, interpretation of data, drafting the article and revising it for submission, and approval of the version to be submitted.

  • Funding This work was supported by grants from the Montreal-Israel Cancer Research Fund, the Gloria Shapiro Fund, the Anne Marie and Mitch Garber Fund, and the Susan and Jon Wener Fund.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval Hospital Institutional Research Board (Protocol #2019-1536).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available upon reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information.